Google is changing, starting from October 9, which AI models are available for Gemini users. The change will affect both users who utilize the service for free and subscribers of the Google AI Plus plan.
According to information disclosed by 9to5Google based on an updated support document from Google, users without a subscription to an AI plan will lose access to the models Gemini 3.6 Flash and Gemini 3.1 Pro. They will be limited to Gemini 3.5 Flash-Lite.
The change occurs after Google changed, in May, the way Gemini usage is billed based on the amount of computational resources used. Now, the company will also start differentiating the available models according to the subscribed plan.
What changes for each plan
- Users who use Gemini without a paid plan will continue to have access to Flash-Lite, but will lose access to the Flash and Pro models;
- For subscribers of the Google AI Plus plan, who pay US$ 4.99 per month (about R$ 27), Gemini will continue to offer Flash-Lite and Flash, but the Pro model will no longer be available;
- O 9to5Google states that these users will receive an email informing when the change will be applied;
- Existing subscribers of the AI Pro and AI Ultra plans will not have a reduction in access to models. Both will continue with Flash-Lite, Flash, and Pro.
Check below a table that clarifies the topic:
| Plan | Flash-Lite | Flash | Pro |
|---|---|---|---|
| No plan | ✓ | — | — |
| AI Plus | ✓ | ✓ | — |
| AI Pro | ✓ | ✓ | ✓ |
| AI Ultra | ✓ | ✓ | ✓ |
Read more:
- Gemini now answers the phone for you — and you can take over at any time
- Gemini gains a face: Google's new AI speaks, reacts, and syncs lips in real-time
- Adobe combines over 80 tools in Claude and brings Photoshop, Lightroom, and Firefly to Gemini
While free users and AI Plus will lose access to models, the AI Pro plan will receive a new resource.
Subscribers to the US$ 19.99 per month (around R$ 109) plan will now have access to Deep Think, an option described by Google as focused on “maximum parallel reasoning”.Until now, the feature was available only to AI Ultra subscribers, whose plans cost US$ 99.99 and US$ 199.99 per month (around R$ 544 and R$ 1,087).
While free users and AI Plus will lose access to models, the AI Pro plan will receive a new resource – Image: Runrun2/Shutterstock
Gemini will have new reasoning levels
Another change planned for October is the inclusion of three effort levels for the models available in the app: low, medium, and high.
The idea is to allow users to choose how much effort Gemini should use to solve a task. According to the description cited by 9to5Google, higher levels may increase the model's ability to perform tasks and produce more detailed answers, but they will also consume more of the usage limit.
Currently, Gemini offers the “Extended thinking” option, aimed at solving complex problems. The new levels should bring the app's functionality closer to the options already available in Google AI Studio and Antigravity.
And what about the Gemini 4 Argon?
It is still not clear where the Gemini 4 Argon, Google's new advanced model, will fit within this tier structure.
Google announced that the model will be initially available to subscribers of AI Ultra, but it has not yet informed whether Argon will be classified as a Pro model or if it will be part of a new category above this level.
The Google itself also maintains support pages that describe usage limits and model availability as subject to changes, depending on factors such as system capacity and experiments.
In practice, starting from October 9th, the main change for those using Gemini without paying will be the loss of access to the Flash and Pro models, while AI Plus subscribers will also lose access to Pro. Higher plans, on the other hand, will maintain access to all three models and gain new reasoning features.
Rodrigo Mozelli is a journalist graduated from the Universidade Metodista de São Paulo (UMESP) and, currently, is a writer at Olhar Digital.
